106 PHILOSOPHICAL TRANSACTIONS. [aNNO 1781. 



The method Mr. T. would recommend, is as follows. A quantity of powder 

 being provided, which, from any previous examination or trial, is known to be 

 of a proper degree of strength to serve as a standard for the proof of other 

 powder, a given charge of it is to be fired, with a fit bullet, in a barrel sus- 

 pended by 2 pendulous rods, and the recoil is to be carefully measured on the 

 ribbon. And this experiment being repeated 3 or 4 times, or oftener if there be 

 any difference in the recoil, the mean and the extremes of the chords may be 

 marked on the ribbon by black lines drawn across it, and the word proof may be 

 written on the middle line; or if the recoil be uniform (which it will be to a suf- 

 ficient degree of accuracy, if care be taken to make the experiments under the 

 same circumstances) then the proof mark is to be made in that part of the ribbon 

 to which it was constantly drawn out by the recoil in the different trials. 



The recoil, with a known charge of standard powder, being thus ascertained 

 and marked on the ribbon, let an equal quantity of any other powder, that is to 

 be proved, be fired in the same barrel, with a bullet of the same weight, and 

 every other circumstance alike ; then if the ribbon is drawn out as far or farther 

 than the proof mark, the powder is as good or better than the standard ; but if 

 it falls short of that distance, it is worse than the standard, and is to be rejected. 

 For the greater the velocity is with which the bullet is impelled, the greater will 

 be the recoil ; and when the recoil is the same, the velocities of the bullets are 

 equal, and the powder is of the same degree of strength, if the quantity of the 

 charge be the same. And if care be taken in proportioning the charge to the 

 weight of the bullet, to come as near as possible to the medium proportion that 

 obtains in practice, the determination of the goodness of gunpowder, from the 

 result of this experiment, cannot fail to hold good in actual service. 



The length of the bore recommended, is 30 inches, and its diameter 1 inch, 

 consequently it is just 30 calibres in length, and will carry a leaden bullet of 

 about 3 ounces. The barrel may be made of gun metal, or of cast iron as that 

 is a cheaper commodity ; but great care must be taken in boring it, to make the 

 cylinder perfectly straight and smooth, as well as to preserve the proper dimen- 

 sions. Of whatever metal the barrel is made, it ought to weigh at least 50 lbs. 

 in order that the velocity of the recoil may not be too great ; and the rods by 

 which it is suspended should be 5 feet in length. The vent may be about ~ of 

 an inch in diameter ; and it should be bouched or lined with gold, in the same 

 manner as the touch-hole is made in the better kind of fowling-pieces, in order 

 that its dimensions may not be increased by repeated firing. 



The bullets should be made to fit the bore with very little windage ; and it 

 would be better if they were all cast in the same mould, and of the same parcel 

 of lead, as in that case their weights and dimensions would be more accurately 

 the same, and the experiments would of course be more conclusive. The stated 





